• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.465-466
• /
• 2022

본 논문에서는 딥러닝 영상인식 기술을 활용한 객체검출 모델인 YOLOv4를 활용하여 차량의 모델과 번호판인식 시스템을 제안한다. 본 논문에서 제안하는 시스템은 실시간 영상처리기술인 YOLOv4를 사용하여 차량모델 인식과 번호판 영역 검출을 하고, CNN(Convolutional Neural Network)알고리즘을 이용하여 번호판의 글자와 숫자를 인식한다. 이러한 방법을 이용한다면 카메라 1대로 차량의 모델 인식과 번호판 인식이 가능하다. 차량모델 인식과 번호판 영역 검출에는 실제 데이터를 사용하였으며, 차량 번호판 문자 인식의 경우 실제 데이터와 가상 데이터를 사용하였다. 차량 모델 인식 정확도는 92.3%, 번호판 검출 98.9%, 번호판 문자 인식 94.2%를 기록하였다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.3
• /
• pp.1-8
• /
• 2016

We propose an adaptive histogram stretching algorithm for application to a car's personal recorder. The algorithm was used for pre-processing to detect the license plate region in an image from a personal recorder. The algorithm employs a Probability Density Function (PDF) and Cumulative Distribution Function (CDF) to analyze the distribution diagram of the images. These two functions are calculated using an image obtained by sampling at a certain pixel interval. The images were subjected to different levels of stretching, and experiments were done on the images to extract their characteristics. The results show that the proposed algorithm provides less deterioration than conventional algorithms. Moreover, contrast is enhanced according to the characteristics of the image. The algorithm could provide better performance than existing algorithms in applications for detecting search regions for license plates.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.8 no.4
• /
• pp.270-274
• /
• 2015

We suggest that HSV Color Space may be used to detect a vehicle detecting system at nighttime. It is essential that a licence plate should be extracted when a vehicle is under surveillance. To do so, a licence plate may be enlarged to certain size after the aimed vehicle is taken picture from a distance by using Pan-Tilt-Zoom Camera. Either Mean-Shift or Optical Flow Algorithm is generally used for the purpose of a vehicle detection and trace, even though those algorithms have tendency to have difficulty in detection and trace a vehicle at night. By utilizing the fact that a headlight or taillight of a vehicle stands out when an input image is converted in to HSV Color Space, we are able to achieve improvement on those algorithms for the vehicle detection and trace. In this paper, we have shown that at night, the suggested method is efficient enough to detect a vehicle 93.9% from the front and 97.7% from the back.

• Journal of the Institute of Convergence Signal Processing
• /
• v.2 no.1
• /
• pp.1-11
• /
• 2001

This paper presents a learning-based approach for the construction of license Plate recognition system. The system consist of three modules. They are respectively, car detection module, license plate recognition module and recognition module. Car detection module detects a car in the given image sequence obtained from the camera with simple color-based approach. Segmentation module extracts the license plate in detect car image using neural network as filters for analyzing the color and texture properties of license plate. Recognition module then reads characters in detected license plate with support vector machine (SVM)-based characters recognizer. The system has been tested from parking lot and tollgate, etc. and have show the following performances on average: Car detect rate 100%, segmentation rate 97.5%, and character recognition rate about 97.2%. Overall system performances is 94.7% and processing time is one sec. Then our propose system does well using real world.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.11
• /
• pp.511-520
• /
• 2016

Automatic License Plate Detection (ALPD) is a key technology for a efficient traffic control. It is used to improve work efficiency in many applications such as toll payment systems and parking and traffic management. Until recently, the hand-crafted features made for image processing are used to detect license plates in most studies. It has the advantage in speed. but can degrade the detection rate with respect to various environmental changes. In this paper, we propose a way to utilize a Faster Region based Convolutional Neural Networks (Faster R-CNN) and a Conventional Convolutional Neural Networks (CNN), which improves the computational speed and is robust against changed environments. The module based on Faster R-CNN is used to detect license plate candidate regions from images and is followed by the module based on CNN to remove False Positives from the candidates. As a result, we achieved a detection rate of 99.94% from images captured under various environments. In addition, the average operating speed is 80ms/image. We implemented a fast and robust Real-Time License Plate Detection System.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.06c
• /
• pp.407-411
• /
• 2007

이 논문은 우리나라 차량 영상에서 번호판 영역을 추출하는 알고리즘을 제안한다. 우리나라 번호판은 하단에 네개의 숫자를 포함하고 있으므로, 네 개의 숫자를 찾으면 번호판을 추출 할 수 있다. 제안하는 방법은 입력된 영상에서 숫자의 가능성을 가진 연결 요소를 검출하고 이들을 군집화 한다. 군집화 된 연결요소들을 바탕으로 숫자 네개(4-digits) 후보를 생성한다. 4-digits 후보들을 인식하여 숫자의 가능성을 측정하고, 적합도로 변환한다. 후보영역 중 적합도가 가장 높은 영역을 번호판 영역으로 추출한다. 적합도는 Perfect Metrics 방법으로 측정하였다. 제안하는 방법을 주간 영상 4600장과 야간 영상 264장으로 테스트 한 결과 각각 97.23%, 95.45%의 검출률과 0.09%, 0.11%의 오검출률을 얻었다.

• Journal of Korea Society of Industrial Information Systems
• /
• v.24 no.2
• /
• pp.47-55
• /
• 2019

In this paper, we propose a method for vehicle plate detection using depth information which is not influenced by illumination. The 3D camera coordinates of pixels in each block are obtained by using the depth information. Factors of the plane in the block are calculated by 3D coordinates of pixels. After that, the plane similarity between adjacent blocks is calculated by comparing between factors of planes. The adjacent blocks are grouped if the plane similarity is high so that the plane areas are detected. The actual height and width of the plane area are calculated by using depth information and compared with the vehicle plate in order to detect the vehicle plate.

• Journal of Broadcast Engineering
• /
• v.26 no.4
• /
• pp.453-462
• /
• 2021

A lot of license plate data is required for car number recognition. License plate data needs to be balanced from past license plates to the latest license plates. However, it is difficult to obtain data from the actual past license plate to the latest ones. In order to solve this problem, a license plate recognition study through deep learning is being conducted by creating a synthetic license plates. Since the synthetic data have differences from real data, and various data augmentation techniques are used to solve these problems. Existing data augmentation simply used methods such as brightness, rotation, affine transformation, blur, and noise. In this paper, we apply a style transformation method that transforms synthetic data into real-world data styles with data augmentation methods. In addition, real license plate data are noisy when it is captured from a distance and under the dark environment. If we simply recognize characters with input data, chances of misrecognition are high. To improve character recognition, in this paper, we applied the DeblurGANv2 method as a quality improvement method for character recognition, increasing the accuracy of license plate recognition. The method of deep learning for license plate detection and license plate number recognition used YOLO-V5. To determine the performance of the synthetic license plate data, we construct a test set by collecting our own secured license plates. License plate detection without style conversion recorded 0.614 mAP. As a result of applying the style transformation, we confirm that the license plate detection performance was improved by recording 0.679mAP. In addition, the successul detection rate without image enhancement was 0.872, and the detection rate was 0.915 after image enhancement, confirming that the performance improved.

• Journal of IKEEE
• /
• v.15 no.4
• /
• pp.274-280
• /
• 2011

The implementation of the recognition system of a vehicle license plate and the Linux OS environment which is built in SoC Embedded system and its test result are presented in this paper. In order to recognize a vehicle license plate, each character has to be extracted from the whole image of a license plate and the extracted image is revised for the template matching. Labeling technique and numerical features are used to detect the vehicle license plate. Each character in the license plate has coordinates. The extracted image is revised by comparison of the numerical coordinates and recognized through template matching method. The experimental results show that the license plate detection rate is 96%, and a character recognition rate is 73%, and a number recognition rate is 97% for about 300 license plate images. The average time of the recognition in the embedded board is 0.66 sec.

• Journal of the Korea Convergence Society
• /
• v.9 no.9
• /
• pp.31-39
• /
• 2018

In this paper, we proposed a license plate detection method based on contour extraction that adapts to environmental changes. The proposed method extracts contour lines using DoG (Difference of Gaussian) to remove unnecessary noise parts in the contour extraction process. Binarization was applied in ugly outline images, and erosion and dilation operations were used to emphasize the contour of the character part. Then, only the outline of the ratio of the characters of the plate was extracted through the ratio of the width and height of the characters. And the case where the outline is the longest is estimated by estimating the characters of the license plate. For the experiment, we applied 130 image data to license plate on the front of the vehicle, oblique environment, and environment images with various backgrounds. I also experimented with motorcycle images of different license plate patterns. Experimental results showed that the detection rate of the oblique image was 93% and that of the various background environment was 70% in the motorcycle image but 98% in the front image.